Chapter 63. Neuraxial Anesthesia in Outpatients

Since the first description of subarachnoid anesthesia by Bier, it has remained the simplest and most effective technique of regional anesthesia. The use of neuraxial techniques in outpatients has been a more recent development, awaiting ready availability of newer needles that reduced the side effect of postdural puncture headache to an acceptable level in the ambulatory setting. The last 10 years have seen a dramatic increase in the use of these techniques in outpatients, for several reasons.

Simplicity and effectiveness should make central neuraxial techniques (either spinal or epidural) ideal in the outpatient setting. Both spinal and epidural anesthesia are more familiar to practitioners than peripheral nerve blocks and are easier to perform because they do not require nerve localization techniques. They can be performed rapidly and without assistance. Neuraxial techniques are effective for lower abdominal, perineal, and lower extremity surgery, and are among the best choices for practitioners who are just starting to incorporate regional anesthetic techniques in an outpatient practice. They also provide optimal outcomes in most of the important aspects of outpatient anesthesia. Patients with neuraxial blocks have lower pain scores on admission to PACU than patients receiving general anesthesia (GA).1–5 Their frequency of postoperative nausea and vomiting (PONV) appears to be at most one-third of that after GA.6 Most importantly, the frequency of phase 1 PACU bypass is high,1,6,7 and discharge times are comparable to even the fastest for GA techniques if appropriate local anesthetic agents and dosages are chosen.8 Modern small-gauge rounded point needles and short-acting drugs have reduced the side effects that were of concern in the past. With propofol infusions available to provide light but transient sedation, the objection to “being awake” during regional techniques has also disappeared, leaving neuraxial techniques as an excellent choice.

This chapter will focus on the advantages, disadvantages, and practical points associated with spinal and epidural anesthesia in the outpatient setting.

Advantages

Spinal Anesthesia

Spinal anesthesia (SA) is one of the simplest and most reliable of regional anesthesia techniques. The external anatomic landmarks are easily identified. The block can be performed with minimal discomfort, the end-point of cerebrospinal fluid flow is unmistakable, and the onset of anesthesia is more rapid than with any other regional technique. Because of the rapidity of onset, the block can be performed in the operating room without the requirement for additional personnel or a block room. The efficient performance of the block does not add substantially to operating room time any more than for the induction of GA. The onset of the sensory blockade is sufficiently rapid to attain surgical anesthesia by the time the positioning and preparation of the patient are completed. A variety of local anesthetic agents are available that can provide a wide range of duration of surgical anesthesia. The risk of nausea can be reduced if systemic opioids are avoided. Likewise, nausea and vomiting2,9,10 and residual somnolence associated with general anesthetics or heavy premedication can be avoided, allowing a rapid return to full alertness in the PACU. SA is also a technique with a high degree of patient familiarity because of its use in obstetrics and thus is more likely to be accepted by many patient populations. In addition, it employs the lowest dose of local anesthetic of any major regional anesthetic technique and has minimal potential for systemic toxicity.

Epidural Anesthesia

Epidural anesthesia (EA) shares many of the advantages of SA, particularly clinician familiarity, simplicity of superficial landmarks, and ease of performance of the block. It has the additional advantage of allowing a continuous catheter to be placed in the epidural space, which creates the potential for tailoring both the segmental spread and duration of the block. Although it is a more flexible technique, this advantage is attained at the price of a slower onset of surgical anesthesia.

Combined Spinal–Epidural Anesthesia

Combined spinal–epidural anesthesia (CSEA) is also a useful technique in the outpatient setting. The procedure is technically more challenging: once the epidural space is identified, the spinal needle must be introduced through the epidural needle and advanced further into the subarachnoid space. After the local anesthetic is injected, the spinal needle is withdrawn and the epidural catheter is inserted into the epidural space and taped in place. This technique requires more time and technical skill, but CSEA provides the advantages of the rapid onset and dense block of SA along with the flexibility of an indwelling catheter to allow incremental and repeated injections to achieve the desired segmental spread and duration of surgical anesthesia. This technique has been used effectively for extracorporeal shock wave lithotripsy procedures, where the duration of treatment may be unpredictable. Another advantage of the combined technique is the rapid onset of dense perineal anesthesia, which may not be provided by lumbar EA alone. It has also been used for knee arthroscopies when low doses of subarachnoid local anesthesia are used to provide a predictable short duration, but may be supplemented by epidural injection of local anesthetic if further spread or duration of blockade are needed.11 Although the technique combines some of the disadvantages of both neuraxial procedures, it also maximizes the advantages and positive aspects of both SA and EA.

Drawbacks

Neuraxial anesthesia does have potential disadvantages. SA is typically a single-injection blockade, and thus careful attention must be paid to selection of the appropriate local anesthetic agent and dose. If the surgical duration was underestimated or becomes prolonged for unexpected reasons, supplemental GA may be needed. The “single-injection” aspect of SA frequently induces clinicians to give “just a little bit more” drug to ensure adequate distribution and duration; however, this tendency must be resisted, as the downside of this increased dosing pattern is a prolonged recovery and discharge time.6

Postdural puncture headache (PDPH) remains a risk with SA. Newer pencil-point, smaller gauge needles have significantly reduced this frequency to less than 3%.12 It is even less frequent in patients older than age 40. Although PDPH does not result in long-term neurologic damage and usually is not a prolonged inconvenience for the patient, it must be acknowledged in the discussion of SA with the outpatient. If it is inconvenient for a patient to return for an epidural blood patch or essential that the patient not have this debility, alternative anesthetic techniques should be considered.

The most recent concerns about SA for outpatients have revolved around potential toxicity of local anesthetic drugs. A major concern was the reporting of permanent neurologic damage associated with very high doses of concentrated lidocaine injected through spinal microcatheters. This has not been a problem with standard doses of the local anesthetics used for single-injection spinal anesthetics, although all anesthetics injected in the subarachnoid space are potentially neurotoxic.13 A more common, relevant concern has been the symptoms of neurologic irritation associated primarily with lidocaine. This syndrome of “transient neurologic symptoms” (TNS) consists of a burning type of back pain radiating into the buttocks or legs that appears 6–24 h after the resolution of SA and can persist for 1 to 6 days.14 TNS occurs approximately 15–30% of the time, with the highest frequency following lidocaine SA.15 Obese outpatients are more susceptible, especially those having procedures performed in the lithotomy or knee arthroscopy positions.16 Although no sensory or motor deficits are associated with this syndrome,17 and to date no persistent neurologic deficits, it is nevertheless a significant source of morbidity in some patients. Many practitioners have sought alternatives to lidocaine to reduce the incidence of TNS (see section on Lidocaine).

EA also has some potential drawbacks in the outpatient setting. Its slower onset of blockade compared with SA may cause a slight delay. However, if induction of EA is performed in a block room outside the operating room, the onset of anesthesia with drugs such as chloroprocaine (2-CP) or lidocaine is so rapid that there is little delay in the onset of surgery, and the use of EA may even promote operating room efficiency.18 Other drawbacks associated with EA include the greater risk for postdural headache if an unintentional dural puncture occurs. Since the potential for headache is directly related to the size of the needle, the use of the larger gauge epidural needles may represent a greater risk, although the incidence of PDPH from an unintentional dural puncture in experienced hands should be less than 0.5%. EA involves the use of larger doses of local anesthetic drugs, and thus represents a greater potential for systemic toxicity than occurs with SA. The careful use of safety steps is just as appropriate for the outpatient as for the inpatient.19

Another limitation of both neuraxial techniques is the absence of residual analgesia. Multiple randomized comparisons of neuraxial techniques to GA show that early pain is significantly less with the regional techniques,1–3,20,21 but once the block has resolved, some alternative mode of analgesia must be provided for the patient. This may be accomplished by the use of local anesthetic infiltrated into the wound, intraarticular injection of local anesthetic, or even a supplemental peripheral nerve block (eg, a femoral nerve or ankle block). The need for additional analgesia may not be an issue after relatively less painful procedures, such as diagnostic knee arthroscopy. Nevertheless, the possibility for breakthrough pain must be considered in the planning of the central neuraxial anesthesia for the outpatient.

A common concern is the potential for difficulty with urination following neuraxial blockade. With higher doses of longer acting local anesthetics, the bladder is distended beyond its normal cystometric capacity during the prolonged duration of neural blockade and may be unable to return to normal function once the sensory blockade dissipates.22 Fortunately, with the short-duration central neuraxial blockades that are usually employed in the outpatient setting, bladder function returns promptly following complete resolution of the blockade. Patients can be successfully discharged home after short-duration spinal anesthetics with procaine, 2-CP, lidocaine, and even low doses of bupivacaine.23 The use of certain additives, such as epinephrine, may impede this recovery.24,25 The requirement for postoperative voiding is not essential with short-acting local anesthetics or low dose (<6 mg) bupivacaine spinal anesthetic techniques.

Hernia Repair

Inguinal herniorrhaphy is one of the most common procedures performed on an outpatient basis. Neuraxial anesthesia provides excellent anesthesia and motor relaxation for this operation.26 SA with procaine or lidocaine in a hyperbaric solution (to give cephalad spread to the T4 through T6 level) is usually sufficient to provide high enough block with appropriate discharge times. The heavier solution (created by the addition of glucose to the local anesthetic) promotes spread to the “lowest” portion of the spinal canal, which is actually the midthoracic region when the patient is in the supine position. Using this technique creates higher spread of the dose, may allow a slightly lower dose to achieve anesthesia,27 and can provide a more rapid resolution of blockade compared with the same dose of local anesthetic injected as an isobaric solution. This more rapid resolution can be negated if higher doses of local anesthetic are used because of insecurity regarding adequate spread of anesthesia. Inappropriately high doses increase the risk of prolonged blockade and delayed discharge.

Although postoperative voiding is not usually a problem after short-acting or low-dose bupivacaine, patients with hernia repairs frequently have urinary retention simply because of the postoperative pain that produces reflex inhibition of the voiding pathway. Generous use of local anesthetic infiltration during and after the procedure may help prevent this, but many centers still require hernia patients to void before discharge regardless of the anesthetic technique. EA offers an alternative for this surgical procedure because of the segmental band of anesthesia and the opportunity to use shorter acting drugs such as 2-CP or lidocaine, but with the ability to administer these agents via the indwelling epidural catheter to ensure an adequate distribution, depth, and duration of surgical anesthesia as dictated by the nature of the surgical procedure.

Anorectal Procedures

Whereas EA may have a delayed onset in the sacral fibers, intrathecal anesthesia limited to specific dermatomal fibers is an ideal application of SA. For lithotomy procedures, a traditional “saddle block” technique with hyperbaric 2-CP or procaine in small doses (20–30 mg) is appropriate. This may add some time to the procedure because of the need to keep the patient in the sitting position for at least 5 min to concentrate the local anesthetic in the sacral area. Because of the risk of TNS, lidocaine may not be an ideal choice. An excellent alternative is the use of the prone jackknife position, where procaine or 2-CP, as well as lidocaine can be made hypobaric.28 This is easily accomplished by adding an equal volume of sterile water to the local anesthetic solution, usually 2 mL of each. This technique is efficient in the operating room because the block can be performed with the patient in the operating position, and onset is fast enough to allow surgery to begin as soon as the surgical preparation is complete. One hazard is that the level of anesthesia may rise if the patient is turned supine with the head elevated after a short procedure. Patients should be recovered form this technique in the full supine position for the first hour. Again, provision for adequate postoperative analgesia should be made by generous injection of local anesthetics during the procedure by the surgeon. The choice of an appropriate short-duration local anesthetic is important to provide competitive discharge times.20

Laparoscopy

Although laparoscopic procedures of the upper abdomen are not tolerated well with regional techniques, pelvic laparoscopy appears to be suitable with neuraxial blockade in some hands. Vaghadia and colleagues have shown the successful use of low-dose SA for outpatient gynecologic laparoscopy. For this situation, lidocaine 25 mg combined with fentanyl 20 mcg injected in the sitting position produces adequate sensory anesthesia with minimal motor blockade, which was well tolerated by their patients.29 This is another example of the use of a hypobaric solution, this time providing upward spread of the solution by the dilution to a total volume of 3 mL with sterile water. In this case, the dilution also creates a lower dose of local anesthetic at each dermatome, and thus allows a less dense block and a more rapid resolution of the dilute anesthetic. The combination of low total dose and dilution provides rapid discharge and a high degree of patient satisfaction. This SA dosing strategy requires a gentle and skilled surgeon capable of performing the operation with less extensive distention of the peritoneum. If such a partnership can be arranged, patients appear to benefit from this technique.

Knee Surgeries

Knee surgeries are excellent situations for using neuraxial techniques, because they provide good surgical conditions and rapid recovery. Epidural blockade with a continuous technique is especially useful for providing a variable length of anesthesia for unpredictable or longer procedures, such as cruciate ligament repairs. With short- or intermediate-duration local anesthetics, timing of dosage can be managed to allow rapid and competitive discharge times.3,4,8,30 The drawback is that larger initial doses are required to provide the lower lumbar spread to anesthetize the knee, especially the posterior portion. With the short-duration drugs, the onset time is usually rapid enough that surgical anesthesia is present by the time the surgical preparation and draping is completed.

SA provides even more rapid onset, and has been used successfully, especially for the short predictable procedures such as diagnostic arthroscopy or partial menisectomy. The challenge here is choosing the correct drug and dose to provide adequate anesthesia but competitive recovery. Lidocaine had been the drug of choice before the TNS controversy and is still used in many institutions and provides competitive early discharge.5 Hodgson has used procaine as an alternative, with lower TNS but other side effects.31 Others have reported success with low-dose bupivacaine spinal block, with a low incidence of TNS.32 This technique requires addition of 20 to 25 mcg of fentanyl to 5 mg bupivacaine to provide sufficient density of anesthesia with a low enough dose to permit rapid resolution. The high variability of spread and duration of bupivacaine, however, makes this a challenging technique.33 Another approach to reduce TNS has been the use of very low dose lidocaine. Ben-David reported a 10-fold decrease in TNS (incidence comparable to bupivacaine) when he used 20 mg of lidocaine with 25 mcg of fentanyl,34 but others have not been able to confirm his results.35 Most recently, the investigation of 2-CP as a spinal anesthetic has provided both rapid discharge and a suggestion of a low incidence of TNS;36 it may therefore be a useful alternative in the future.

Another variation of spinal technique is unilateral spinal blockade. This approach uses a hyperbaric solution, keeping the patient in the lateral decubitus position for 10 to 15 min after injection to allow concentration of the local anesthetic in the dependent nerve roots. This reduces the usual spread of the drug to the midthoracic level and allows smaller doses and less sympathetic blockade, and thus a shorter duration with fewer side effects. This technique has been employed with low-dose bupivacaine, ropivacaine, and levobupivacaine37,38 and has been shown to have advantages compared with conventional hyperbaric techniques and to GA.2,39–42 The main disadvantage of this technique is the longer time for induction of anesthesia, but this may be overcome if the block can be performed in a holding area while the operating room is being prepared.

Most diagnostic and minor knee procedures do not include significant postoperative pain, and thus the patient is usually free to leave when the neuraxial anesthetic resolves. With the more painful procedures, such as patellar tendon cruciate ligament repair, the use of supplemental peripheral nerve blocks can provide significant analgesia and speedy discharge.43,44

Foot Surgery

Foot surgery requires more sacral anesthesia, and thus EA may not be as useful. Spinal blockade has been used successfully, including the unilateral techniques described previously.45 The major problem with foot surgery is that it can be extremely painful, and SA provides no residual analgesia when the block resolves. This is an excellent opportunity for the combination of a dense spinal blockade for the surgery, followed by a long-acting peripheral nerve blockade, or even the insertion of a continuous peripheral nerve catheter on the sciatic nerve to provide postoperative analgesia.46

Chloroprocaine

For epidural injection, 2-CP has been the standard drug for the outpatient setting because of its short duration (complete resolution within 2 h, with a narrow variation [Figure 63–1]). 2-CP provides discharge times that are competitive with short-acting GA techniques.47,48 The onset of blockade is sufficiently rapid that if the block is performed in a preoperative holding area, the patient usually has obtained surgical anesthesia during the time of transition to the operating room and patient positioning, preparation, and draping. The major drawback with 2-CP as the epidural agent has been the issue of back pain associated with this drug. Back pain appears to be dose-related.49 The use of single-injection techniques, limited to less than 25 mL, decreases the incidence of back pain. The introduction of preservative-free preparations also appears to have eliminated the risk of neurotoxicity that was previously associated with this drug when unintentional high-dose subarachnoid injection was performed.

2-CP had originally been introduced for SA in the 1950s,50 and the advent of the new preservative-free solutions has allowed its reexamination for subarachnoid use. The current 2% and 3% preservative-free preparations are slightly hyperbaric at 37°C and reliably produce sensory block (to pinprick) to the T8 through T4 level when used in a dose range of 40 to 60 mg.51,52 Kouri and Kopacz demonstrated in a volunteer model that 40 mg of 2-CP provides tolerance to simulated surgical stimulus for up to 60 min at the T12 dermatome, which was equivalent to anesthesia provided by 40 mg of 2% lidocaine (Figure 63–2).53 More importantly, 40 mg 2-CP resulted in complete resolution of sensory block and the ability to spontaneously void (103 min) sooner than with equipotent doses of lidocaine (126 min), procaine (151 min),54 or low-dose bupivacaine (191 min).55 The duration of surgical anesthesia can be reliably prolonged by 20 min by the addition of 20 mcg of fentanyl.

Concern has been raised about potential neurotoxicity of 2-CP in an animal model in extremely high doses (equivalent to 1000 mg in a 70-kg man),56 but this level of toxicity does not appear to be any greater than that seen in the same model with lidocaine or prilocaine.57 Although further investigation will be required to confirm that this drug is safe and effective for clinical use, initial experience has not identified a problem of TNS36 and 2-CP may prove to be a suitable alternative to lidocaine.

Procaine

An older alternative for short-duration SA has been the aminoester drug procaine. This drug is available in a commercial 10% solution, which should be diluted by at least 50% before injection. Procaine can be made hyperbaric, isobaric, or hypobaric by the addition of appropriate additives. It has been used in doses of 75 to 100 mg for procedures such as knee arthroscopy and hernia repair with good results, with surgical duration and discharge times equivalent to those found with 50 mg of lidocaine,8,31 but longer than those for 2-CP.54 The addition of fentanyl to procaine appears to produce a higher frequency and severity of itching than is found with the other local anesthetics.58 Nevertheless, procaine is a useful alternative to lidocaine in the United States today because of its lower incidence of TNS than with lidocaine.15 Procaine is not reliable as an epidural anesthetic.

Lidocaine

The traditional short-acting local anesthetic for outpatient SA has been lidocaine. In dosages in the 50-mg range, this local anesthetic provides 60–90 min of surgical anesthesia, with a predictable resolution within 2.5 h, which allows for an acceptable discharge home after most outpatient procedures on the lower abdomen and lower extremity. For patients having procedures not in the lithotomy or arthroscopy positions, the incidence of TNS is <15% and is relatively mild, making lidocaine still an acceptable choice of procedures in the supine position.14 It can also be diluted with sterile water and injected as a hypobaric solution in the prone jackknife position in a 40-mg dose, to provide excellent surgical anesthesia for most anorectal procedures.28

Although diluting the concentration of lidocaine does not reduce the frequency of TNS,59 there has been enthusiasm for the use of small doses of lidocaine potentiated by small doses of fentanyl to provide anesthesia for lithotomy or knee procedures. Vaghadia and colleagues have used 20 mg of lidocaine with fentanyl for SA for gynecologic laparoscopic surgery and have reported no instances of TNS in several of their case series.10,60,61 Ben-David studied a group at high-risk group for TNS (knee arthroscopy patients) and reported an incidence of only 3% TNS with a 20-mg dose of lidocaine potentiated by 25 mcg of fentanyl.62 Other studies have found an incidence of 12% of TNS with 25 mg of lidocaine plus fentanyl in this high-risk population.35 Although many practitioners have changed to using a low dose of lidocaine, many have looked for other alternatives such as procaine and 2-CP.

Lidocaine has also been used for EA in the outpatient setting. Its duration of action is approximately 30 min longer than that of 2-CP,47 and thus epidural lidocaine may be more suitable for surgical procedures in the 60- to 90-min range.

Mepivacaine

Mepivacaine is an aminoamide local anesthetic similar in structure to lidocaine, with a slightly longer duration of action. Mepivacaine has been used successfully for SA for knee arthroscopies in 40-mg doses, with discharge time slightly longer than for lidocaine. There are conflicting reports about the frequency of TNS associated with this drug, with some authors claiming a reduced incidence, and others finding an equivalent frequency of TNS following mepivacaine (vs lidocaine).15

Prilocaine

Although prilocaine is not available in the United States as a local anesthetic for spinal use, it has produced effective SA in clinical trials and a lower incidence of TNS than seen with lidocaine.

Bupivacaine

Bupivacaine is a more potent and longer acting aminoamide local anesthetic and has received interest as a potential alternative to lidocaine for outpatient SA because of the lower incidence of TNS. Unfortunately, in addition to a longer duration, bupivacaine also has a wider variability in the duration of anesthetic action. Liu and coworkers showed in a series of volunteers that each additional milligram of bupivacaine added to SA provides approximately an average 10-min prolongation of surgical anesthesia, but a 20-min prolongation of discharge time.33 More importantly, the standard deviation of the readiness for discharge time was almost an hour, which can translate into unacceptably long discharge times for some outpatients when doses greater than 6 mg are used. This wide standard deviation has also been confirmed at low doses.32 The variability in the height of the block as well as the duration make bupivacaine not easily predictable for outpatient surgical procedures. Nevertheless, many practitioners are willing to accept this variability and potentially prolonged block in exchange for the very low rate of TNS associated with this drug, which is at 3% or less in clinical trials.

Ropivacaine and Levobupivacaine

The aminoamide ropivacaine has been studied for outpatient anesthesia. Although it is effective as a subarachnoid local anesthetic, it appears to be less potent than bupivacaine by approximately 50%.63–65 Although this drug has enjoyed popularity because of its reduced risk of cardiac toxicity in large doses and less motor blockade when used on peripheral nerve blockade, these two advantages are not clinically relevant in its use in a subarachnoid space. Levobupivacaine is another less cardiotoxic enantiomer equivalent to bupivacaine studied for outpatient use,66 but its additional cardiac safety is not a relevant factor in SA. None of the longer acting aminoamides are useful for EA in the outpatient.

Adjuvants

Additives have been used with SA over the years to prolong the duration or to intensify blockade, and thus allow a lower dose of the local anesthetic drug itself.

Changing the baricity of a subarachnoid local anesthetic solution by the addition of either sterile water or dextrose can allow greater control of the spread of local anesthetic to specific areas within the intrathecal space, as described in previous sections. Such manipulations can produce “unilateral” SA, allowing lower doses of local anesthetic and faster same-day discharge. The block frequently crosses over to the opposite side when the patient is finally turned supine, although it is less dense and of shorter duration on the opposite side. The use of such a lower dose allows for more rapid recovery and less chance of urinary retention, but the additional time required to perform this unilateral technique presents a challenge in a rapid-turnover ambulatory surgical setting. In this context, the RA induction room is of tremendous benefit for the safe and efficient placement of unilateral hyperbaric SA for patients undergoing lower extremity surgery.

Epinephrine

Epinephrine was, but is no longer, the classic drug additive for SA. Epinephrine provides prolongation and potentiation for all local anesthetics. It has been shown to be safe in standard doses of 100 to 200 mcg, but it is associated with a significant increase in the time to voiding in patients receiving both bupivacaine and lidocaine.22,24 Epinephrine also produces an unusual flu-like syndrome when added to 2-CP.67 In general, epinephrine does not appear to be a useful additive for outpatient SA unless used in extremely low doses68 and is probably best avoided altogether.

Fentanyl

Fentanyl has been used extensively as an additive in outpatient SA. Fentanyl prolongs and intensifies the block produced by lidocaine and 2-CP without a delay in urination. Thus, intrathecal fentanyl in doses of 20 to 25 mcg may be a useful adjunct for ambulatory SA.25,32,68,69 Unfortunately, fentanyl is associated with pruritus: 100% of volunteers receiving lidocaine or 2-CP SA with fentanyl experienced itching. In the clinical setting, where other factors such as surgical pain and systemic sedatives and analgesics are added, the itching is reported less frequently by patients. Patients receiving lidocaine SA with fentanyl appear to have an incidence of about 25% of mild pruritus. When fentanyl is added to bupivacaine, the incidence appears to double, but again the symptoms remain mild. When fentanyl is added to procaine, itching becomes more noticeable and unpleasant and may even require systemic treatment.58 The itching usually responds to small doses of nalbuphine or diphenhydramine. Nevertheless, the use of fentanyl is effective in reducing the total dose of local anesthetic and allowing a shorter discharge time.32 The longer acting intrathecal hydrophilic opioids such as morphine provide extended postoperative analgesia for monitored inpatients. However, morphine's slow onset of action and low risk of respiratory depression make it an undesirable intrathecal drug in the outpatient setting.

Clonidine

Clonidine is an α2-agonist that potentiates local anesthetics in the subarachnoid space and on peripheral nerves. Doses of 15 to 45 mcg are effective without side effects, but larger doses may produce bradycardia and sedation.70,71 The addition of 15 mcg to 2-CP SA, for example, prolongs motor and sensory block without systemic side effects.71 Unfortunately, in the United States the drug is marketed in relatively expensive large-dose vials, in contrast to the European market where it is available in more practical smaller incremental dosages. For that reason, it has not attained great popularity in the United States, but is worth considering for potentiation of SA. Another additive that has been tested is neostigmine, but it is associated with severe intractable nausea, which makes it unacceptable as an additive at the present time.72

Although peripheral nerve blocks can provide excellent postoperative analgesia if patients are allowed to be discharged before resolution of the block, neuraxial techniques generally require that the patient's block be completely resolved before discharge in order to allow successful ambulation. Spinal and EA regress in slightly different patterns. With lumbar EA, the regression is generally both upward and downward from the points of maximum caudad and cephalad spread, and thus results in the final resolution of sensory anesthesia usually occurring somewhere in the lumbar dermatomes near the level of the original injection. Because of this, patients with epidurals are generally able to ambulate and void as soon as the sensory block is resolved on the thigh. On the other hand, SA tends to resolve in a cephalad-to-caudad fashion, with disappearance of sensory and motor block in the sacral dermatomes occurring last. There has been concern in the past about residual sympathetic blockade with subarachnoid anesthesia even after the sensory block is resolved. It has been shown that if perianal sensation has returned, along with proprioception in the big toe, sympathetic activity is generally back to normal and orthostatic hypotension is not a risk when these patients are ambulated.73 Thus, the resolution of neuraxial blockade can be fairly easily determined with either technique.

Two major issues with recovery are the timing of discharge and postoperative analgesia. As mentioned earlier, careful attention must be paid to the selection of the drug and dose in order to allow a rapid resolution of the block at the appropriate time following the surgery. This requires a familiarity with the procedure and the surgeon. More importantly, when the blockade is resolved, some alternative form of analgesia must be provided. The best choice is the use of local infiltration with long-acting local anesthetics by the surgeon at the time of the operation. Other alternatives include injection of peripheral nerve blocks in the PACU as the spinal or epidural is resolving. Examples of this are the use of bupivacaine for ankle blocks after foot surgery or for femoral nerve blocks following anterior cruciate ligament repair. The next best alternative is to rely on nonopioid analgesia for patients who will have mild postoperative pain, such as the use of ketorolac or other nonsteroidal antiinflammatory drugs for patients having mild pain associated with diagnostic knee arthroscopy. If patients are having more severe pain, the use of more potent oral opioids may be required, but this may negate the advantages of regional anesthesia because of the potential for nausea, pruritus, or sedation when these drugs are administered in the PACU.

Discharge Criteria

In general, the discharge criteria, including alertness, hemodynamic stability, and freedom from nausea and vomiting, are the same for patients receiving neuraxial blockade on an outpatient basis as for all other outpatients. The need for an accompanying adult and overnight supervision is also the same. The two central special features of neuraxial blockade patients are the issue of postoperative analgesia and of voiding.

As mentioned previously, postoperative analgesia, in the form of supplemental local anesthesia or systemic analgesic techniques, is necessary. These analgesics need to be tailored to the individual patient. If local anesthesia or a peripheral nerve block is used, they need to be placed early enough to allow adequate analgesia as the spinal or epidural anesthetic wears off. Occasionally, the resolution of neuraxial block reveals that the previous management plan was not adequate, and a subsequent reinjection of local anesthetic or peripheral nerve block (or adjustment of medications) is needed. This may delay discharge, but generally most outpatient anesthesiologists have developed a good sense of the appropriate regimen for patients in these situations.

The issue of voiding is a controversial topic. For outpatients, the use of short-acting neuraxial techniques (SA with 2-CP, procaine, lidocaine, or low-dose bupivacaine or EA with 2-CP or lidocaine) is not associated with overdistention of the bladder. For patients without known risk factors for postoperative urinary retention (PUR), it does not appear necessary to require them to demonstrate an ability to void after resolution of a short-acting neuraxial block. There are several exceptions to this in high-risk groups. This may include patients with previous history of urinary retention, elderly men with prostate symptoms or likelihood of prostate symptoms, or specific surgical procedures that place them at risk for PUR. Those procedures are generally associated with pelvic or groin surgeries such as hernia repair or perirectal procedures after which voiding may be impaired by the pain produced by lower abdominal straining. In general, these patients are regarded as high risk and are usually required to void before discharge in most ambulatory surgery settings. Patients having urologic surgery are also frequently required to void by their surgeon in order to demonstrate absence of bleeding and return of normal bladder function. Other than these exceptions, most patients with neuraxial blockade can be discharged with the same criteria as patients receiving GA. If there is any concern about bladder distension, the use of a bladder ultrasound can easily identify distended bladders and lead to a prophylactic drainage before discharge. Such ultrasound units are frequently available in ambulatory surgery centers.

Patient Instructions

Postoperative instructions for patients receiving neuraxial blockade are the same as for other outpatients. If a supplemental peripheral nerve block has been given, the patients need to be instructed to protect the insensate extremity until the numbness wears off. Nausea and vomiting should be far less common at discharge than with GA techniques. Patients who have not voided before discharge need to be carefully instructed to report back to an emergency room if they are unsuccessful in voiding within 8 h after discharge. They may be one of the rare patients with voiding difficulties who will need subsequent catheterization to relieve an overdistended bladder. But as mentioned earlier, this is no indication for delaying discharge in the low-risk group.

Problems that are specific to neuraxial blockade include the potential for PDPH and neurologic symptoms following resolution of the block. The first of these is the most common, yet should still occur at less than 1% frequency with the appropriate equipment and patient selection. Nevertheless, all patients receiving spinals (or unintentional dural puncture during EA) need to be given a specific telephone number to call for follow-up and a detailed description of the potential symptoms of a PDPH. They need to be advised about the potential need to return to the hospital for an epidural blood patch if severe symptoms do develop. Questioning for symptoms of such a headache needs to be part of the routine phone call follow-up that is performed in outpatient units. If a headache does develop, conservative therapy at home is usually the most appropriate. If it persists for several days or is severe enough to interfere with daily function of the patient, the performance of an epidural blood patch on an outpatient basis is usually sufficient to cure the symptoms.

Neurologic symptoms are also possible and need to be evaluated in the follow-up phone call. The most common is the description of TNS following SA. Although it occurs more frequently with lidocaine, it also has been reported after every local anesthetic, so it should be searched for in every postspinal anesthetic patient. The symptoms may occur 6–24 h following discharge and traditionally are back pain radiating to the buttocks or legs. Fortunately, most of the symptoms are mild and transient. Although NSAIDs have been recommended for the therapy, there is really no effective treatment at the current time. The patient should be reassured that these symptoms usually regress spontaneously within 48 h, and that they do not represent any permanent neurologic deficit, nor is there any expectation that they will recur again if the patient has another spinal in the future.

A much less frequent complication, but still a possibility in the outpatient, is the development of nerve compression symptoms related to epidural hematoma formation. Although this symptom is primarily a phenomenon of inpatients receiving anticoagulants, it has been reported in the outpatient setting.74 The unique feature of this syndrome is the development of back pain associated with motor weakness. This is distinctly different from the TNS syndrome, which does not include any motor symptoms. This syndrome may also develop immediately after discharge. Presentation of any motor weakness mandates an immediate return to the hospital and evaluation of the patient with a CAT scan or (preferably) an MRI to eliminate the possibility of epidural clot formation.

Overall, neuraxial techniques have an important role in the outpatient setting. They are simple to perform and highly reliable and are generally associated with a faster onset of action than peripheral nerve blockade discussed elsewhere. They carry the disadvantage that the block must be resolved before patient discharge, and thus other methods of postoperative analgesia must be planned to provide for pain relief on discharge. It is critical to select the dose and drug that is most appropriate for the specific surgical procedure in order to provide an appropriate duration of blockade to allow a timely discharge from the outpatient unit. Nevertheless, with these cautions and appropriate choices, neuraxial blockade has a strong role to play in the outpatient surgical setting.